UFC 15: Collision Course: Predictions & Analysis
UFC 15: Collision Course lands on Friday, October 17, 1997 in Bay St. Louis, Mississippi, USA with 7 bouts on the card. Below is our fight-by-fight breakdown, combining Elo ratings, rolling statistical trends, style matchup data, and betting market context into a pick for every bout.
Quick Picks
| Matchup | Pick | Confidence | Prob |
|---|---|---|---|
| Maurice Smith vs David AbbottHeavyweight | David Abbott | Toss-up | 52% |
| Mark Kerr vs Dwayne CasonHeavyweight | Mark Kerr | Strong | 90% |
| Randy Couture vs Vitor BelfortHeavyweight | Randy Couture | Toss-up | 55% |
| Dave Beneteau vs Carlos BarretoHeavyweight | Dave Beneteau | Strong | 87% |
| Mark Kerr vs Greg StottHeavyweight | Mark Kerr | Strong | 94% |
| Dwayne Cason vs Houston DorrHeavyweight | Dwayne Cason | Strong | 81% |
| Alex Hunter vs Harry MoskowitzHeavyweight | Alex Hunter | Strong | 85% |
Like these picks? Bet on DraftKings
Place your bets on the fights above at DraftKings Sportsbook
Bet Now on DraftKingsFight-by-Fight Breakdown
Maurice Smith vs David Abbott
The Heavyweight matchup features Maurice Smith (4-3) taking on David Abbott (8-10).
Smith is rated at 1186 — 298 points above Abbott's 888. Gaps this large usually mean one fighter has been consistently beating better opponents.
Both fighters land in our "Knockout Artist" archetype — precision strikers who sit back, pick their shots, and carry fight-ending power. When mirror matchups like this happen, the edge usually goes to whoever can impose their preferred pace and range.
A few statistical edges stand out. Abbott throws significantly more leather — a 0.1 sig. strike per minute gap. Abbott is far more active with takedowns, averaging 3.6 more per 15 minutes. Smith has tighter striking defense, making opponents miss more often.
The Pick: David Abbott over Maurice Smith. This is essentially a pick'em. The model nudges toward Abbott at 52%, but there's almost nothing separating these two.
Mark Kerr vs Dwayne Cason
The Heavyweight matchup features Mark Kerr (4-0) taking on Dwayne Cason (1-1).
Kerr is rated at 1317 — 379 points above Cason's 938. Gaps this large usually mean one fighter has been consistently beating better opponents.
A few statistical edges stand out. Kerr throws significantly more leather — a 4.3 sig. strike per minute gap. Kerr is far more active with takedowns, averaging 7.5 more per 15 minutes. Cason has tighter striking defense, making opponents miss more often.
The Pick: Mark Kerr over Dwayne Cason. The model is firm on this one: Kerr at 90%.
Randy Couture vs Vitor Belfort
The Heavyweight matchup features Randy Couture (16-8) taking on Vitor Belfort (15-10).
Belfort carries a modest Elo edge (1440 to 1390), the kind of gap that reflects a slightly better run of form rather than a talent chasm. Couture rides a 3-fight win streak into this one.
Stylistically this is Couture's wrestler game against Belfort's knockout artist approach. Couture looks to close distance and put the fight on the mat, while Belfort is patient on the feet, timing counters and loading up when he sees openings. Historically these archetypes are dead-even when they collide.
A few statistical edges stand out. Belfort throws significantly more leather — a 10.2 sig. strike per minute gap. Couture is far more active with takedowns, averaging 2.0 more per 15 minutes. Couture has tighter striking defense, making opponents miss more often.
The Pick: Randy Couture over Vitor Belfort. This is essentially a pick'em. The model nudges toward Couture at 55%, but there's almost nothing separating these two.
Dave Beneteau vs Carlos Barreto
The Heavyweight matchup features Dave Beneteau (3-3) taking on Carlos Barreto (0-1).
Beneteau is rated at 1140 — 242 points above Barreto's 899. Gaps this large usually mean one fighter has been consistently beating better opponents.
A few statistical edges stand out. Beneteau throws significantly more leather — a 3.7 sig. strike per minute gap. Beneteau is far more active with takedowns, averaging 5.8 more per 15 minutes. Barreto has tighter striking defense, making opponents miss more often.
The Pick: Dave Beneteau over Carlos Barreto. The model is firm on this one: Beneteau at 87%.
Mark Kerr vs Greg Stott
The Heavyweight matchup features Mark Kerr (4-0) taking on Greg Stott (0-1).
Kerr is rated at 1317 — 370 points above Stott's 947. Gaps this large usually mean one fighter has been consistently beating better opponents.
A few statistical edges stand out. Kerr throws significantly more leather — a 4.5 sig. strike per minute gap. Kerr is far more active with takedowns, averaging 9.2 more per 15 minutes. Stott has tighter striking defense, making opponents miss more often.
The Pick: Mark Kerr over Greg Stott. The model is firm on this one: Kerr at 94%.
Dwayne Cason vs Houston Dorr
The Heavyweight matchup features Dwayne Cason (1-1) taking on Houston Dorr (0-1).
Cason is rated at 938 — 163 points above Dorr's 775. Gaps this large usually mean one fighter has been consistently beating better opponents.
A few statistical edges stand out. Dorr throws significantly more leather — a 0.0 sig. strike per minute gap. Dorr is far more active with takedowns, averaging 0.0 more per 15 minutes. Dorr has tighter striking defense, making opponents miss more often.
The Pick: Dwayne Cason over Houston Dorr. The model is firm on this one: Cason at 81%.
Alex Hunter vs Harry Moskowitz
The Heavyweight matchup features Alex Hunter (2-0) taking on Harry Moskowitz (0-2).
Hunter is rated at 1164 — 341 points above Moskowitz's 823. Gaps this large usually mean one fighter has been consistently beating better opponents.
A few statistical edges stand out. Hunter throws significantly more leather — a 2.8 sig. strike per minute gap. Hunter is far more active with takedowns, averaging 6.0 more per 15 minutes. Moskowitz has tighter striking defense, making opponents miss more often.
The Pick: Alex Hunter over Harry Moskowitz. The model is firm on this one: Hunter at 85%.
Methodology
Predictions are generated by our ensemble model combining LightGBM (65%) and CatBoost (35%), trained on every UFC fight since 1994. The model uses 23 features including Elo ratings, rolling 5-fight statistical averages, style matchup history, physical attributes, and market odds when available.
On our held-out test set (402 fights from January-September 2023), the model achieves 63.4% accuracy with a log-loss of 0.626. High-confidence picks (>75% probability) hit at 82.7%. For full model transparency, visit our Model page.